All of lore.kernel.org
 help / color / mirror / Atom feed
From: Xiong Zhou <xzhou@redhat.com>
To: Eryu Guan <eguan@redhat.com>
Cc: Xiong Zhou <xzhou@redhat.com>,
	fstests@vger.kernel.org, ross.zwisler@linux.intel.com,
	dan.j.williams@intel.com, jmoyer@redhat.com
Subject: Re: [PATCH v2 2/3] DAX-DIO: skip DAX to non-DAX if unsupported
Date: Tue, 11 Apr 2017 22:26:36 +0800	[thread overview]
Message-ID: <20170411142636.au3fnupvd4imuvlx@XZHOUW.usersys.redhat.com> (raw)
In-Reply-To: <20170411114410.GJ22845@eguan.usersys.redhat.com>

On Tue, Apr 11, 2017 at 07:44:10PM +0800, Eryu Guan wrote:
> On Mon, Apr 10, 2017 at 02:05:52PM +0800, Xiong Zhou wrote:
> > By tweaking SCRATCH_DEV by ndctl, only run DAX mapped
> > area DIO to non-DAX area tests when pmem device is in
> > "memory mode".  > > 
> > Signed-off-by: Xiong Zhou <xzhou@redhat.com>
> > ---
> > 
> > v2:
> >   umount/check scratch fs after test
> >   restore SCRATCH_DEV mode in cleanup
> >   fix variable names
> > 
> >  tests/generic/413 | 26 +++++++++++++++++++++++++-
> >  tests/xfs/260     | 31 ++++++++++++++++++++++++++++++-
> >  2 files changed, 55 insertions(+), 2 deletions(-)
> > 
> > diff --git a/tests/generic/413 b/tests/generic/413
> > index a1cc514..fbe8f11 100755
> > --- a/tests/generic/413
> > +++ b/tests/generic/413
> > @@ -34,6 +34,9 @@ _cleanup()
> >  {
> >  	cd /
> >  	rm -f $tmp.*
> > +	# restore SCRATCH_DEV to original mode
> > +	$NDCTL_PROG create-namespace -f -e $scratch_dev_namespace \
> > +	  -m $scratch_dev_mode > /dev/null 2>&1
> >  }
> >  
> >  # get standard environment, filters and checks
> > @@ -47,9 +50,13 @@ _supported_fs generic
> >  _supported_os Linux
> >  _require_test
> >  _require_scratch_dax
> > +# fsck manually after tests as we will reconfig SCRATCH_DEV
> > +_require_scratch_nocheck
> >  _require_test_program "feature"
> >  _require_test_program "t_mmap_dio"
> >  _require_xfs_io_command "falloc"
> > +_require_ndctl
> > +_require_jq
> >  
> >  prep_files()
> >  {
> > @@ -110,7 +117,7 @@ t_both_nondax()
> >  t_mmap_dio_dax()
> >  {
> >  	t_both_dax $1
> > -	t_dax_to_nondax $1
> > +	[ $skip_dax_to_nondax -eq 0 ] && t_dax_to_nondax $1
> >  	t_nondax_to_dax $1
> >  	t_both_nondax $1
> >  }
> > @@ -126,6 +133,21 @@ do_tests()
> >  	t_mmap_dio_dax $((64 * 1024 * 1024))
> >  }
> >  
> > +skip_dax_to_nondax=0
> > +# config SCRATCH_DEV to memory mode to support DAX mapped
> > +# area DIO to non-DAX area.
> > +
> > +# save original mode
> > +scratch_dev_mode=$(_ndctl_get_pmem_key_value $SCRATCH_DEV mode)
> > +
> > +# get its namespace
> > +scratch_dev_namespace=$(_ndctl_get_pmem_key_value $SCRATCH_DEV dev)
> > +
> > +# skip dax to non-dax dio if config fails
> > +$NDCTL_PROG create-namespace -f -e $scratch_dev_namespace \
> > +  -m memory > /dev/null 2>&1 \
> > +  || skip_dax_to_nondax=1
> > +
> 
> I'm not sure if it's proper to change the pmem device mode in tests.
> IMO, fstests just takes use of user specified TEST|SCRATCH_DEV and
> checks if they meet our requirements.
> 
> So I think the correct way is just check the mode of pmem device and
> _notrun if the mode isn't compatible with this test.

I'm working on splitting this case into 2 cases. Running dax_to_nondax
solely, _notrun if not compatible. Since only this sub case really need
pmem device in specific mode.

Thanks,
Xiong

> 
> Thanks,
> Eryu

  parent reply	other threads:[~2017-04-11 14:26 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-07  7:56 [PATCH 1/3] common: helper to get value from ndctl list by key Xiong Zhou
2017-04-07  7:56 ` [PATCH 2/3] DAX-DIO: skip DAX to non-DAX if unsupported Xiong Zhou
2017-04-12  3:30   ` Ross Zwisler
2017-04-07  7:56 ` [PATCH 3/3] DAX: mmap write readonly file Xiong Zhou
2017-04-07 17:16 ` [PATCH 1/3] common: helper to get value from ndctl list by key Ross Zwisler
2017-04-10  6:05   ` [PATCH v2 " Xiong Zhou
2017-04-10  6:05     ` [PATCH v2 2/3] DAX-DIO: skip DAX to non-DAX if unsupported Xiong Zhou
2017-04-11 11:44       ` Eryu Guan
2017-04-11 13:54         ` Jeff Moyer
2017-04-11 14:26         ` Xiong Zhou [this message]
2017-04-10  6:05     ` [PATCH v2 3/3] DAX: mmap write readonly file Xiong Zhou
2017-04-11 11:46       ` Eryu Guan
2017-04-11 13:56         ` Jeff Moyer
2017-04-11 18:52           ` Ross Zwisler
2017-04-11 22:45             ` Ross Zwisler
2017-04-12  2:52               ` Xiong Zhou
2017-04-12  4:03       ` Ross Zwisler
2017-04-12  6:26         ` Xiong Zhou
2017-04-12 14:46           ` [PATCH v3 0/4] split DAX mmap DIO cases Xiong Zhou
2017-04-12 14:46             ` [PATCH v3 1/4] DAX-DIO: make dax_to_non_dax dio test solo Xiong Zhou
2017-04-12 14:46             ` [PATCH v3 2/4] generic: test mmap io fom DAX to non-DAX Xiong Zhou
2017-04-13  4:11               ` Xiong Zhou
2017-04-13 13:36                 ` Dan Williams
2017-04-14 10:01                   ` Xiong Zhou
2017-04-14 14:49                     ` Dan Williams
2017-04-14 15:22                       ` Ross Zwisler
2017-04-14 15:33                         ` Dan Williams
2017-04-14 15:51                           ` Ross Zwisler
2017-04-12 14:46             ` [PATCH v3 3/4] xfs: test per-inode DAX flag " Xiong Zhou
2017-04-12 14:46             ` [PATCH v3 4/4] generic: mmap write readonly DAX file Xiong Zhou
2017-04-17  7:14             ` [PATCH v4 0/4] split DAX mmap DIO cases Xiong Zhou
2017-04-17  7:14               ` [PATCH v4 1/4] DAX-DIO: make dax_to_non_dax dio test solo Xiong Zhou
2017-04-18 16:31                 ` Ross Zwisler
2017-04-17  7:14               ` [PATCH v4 2/4] generic: test mmap io fom DAX to non-DAX Xiong Zhou
2017-04-17 14:14                 ` Dan Williams
2017-04-17 22:54                   ` Dan Williams
2017-04-17 23:39                     ` Ross Zwisler
2017-04-17 23:47                       ` Dan Williams
2017-04-19  8:40                   ` Xiong Zhou
2017-04-19 15:53                     ` Dan Williams
2017-04-18 10:12                 ` Christoph Hellwig
2017-04-18 14:49                   ` Xiong Zhou
2017-07-28 14:55                   ` Jeff Moyer
2017-04-18 16:32                 ` Ross Zwisler
2017-04-17  7:14               ` [PATCH v4 3/4] xfs: test per-inode DAX flag " Xiong Zhou
2017-04-18 16:36                 ` Ross Zwisler
2017-04-17  7:14               ` [PATCH v4 4/4] generic: mmap write readonly DAX file Xiong Zhou
2017-04-18 17:05                 ` Ross Zwisler
2017-04-19  2:54                   ` Xiong Zhou
2017-04-24 16:34                 ` Ross Zwisler
2017-09-25  8:40               ` [PATCH v5 0/3] fix dax to nondax dio fake failures Xiong Zhou
2017-09-25  8:40                 ` [PATCH v5 1/3] generic: mmap write readonly DAX file Xiong Zhou
2017-09-27  9:57                   ` Eryu Guan
2017-09-25  8:40                 ` [PATCH v5 2/3] generic/413: skip dax to nondax dio test if needed Xiong Zhou
2017-09-27  9:50                   ` Eryu Guan
2017-09-27 15:01                     ` Jeff Moyer
2017-09-25  8:40                 ` [PATCH v5 3/3] xfs/260: remove per-inode DAX flag test Xiong Zhou
2017-09-25 16:59                   ` Ross Zwisler
2017-09-26  0:51                     ` Xiong Zhou
2017-09-28  7:41                 ` [PATCH v6 0/4] fix dax to nondax dio fake failures Xiong Zhou
2017-09-28  7:41                   ` [PATCH v6 1/4] tests: add new group dax Xiong Zhou
2017-09-28  7:41                   ` [PATCH v6 2/4] gitignore: sort src/ binaries name Xiong Zhou
2017-09-28 11:10                     ` Dave Chinner
2017-09-28 12:28                       ` Xiong Zhou
2017-09-28 21:43                         ` Dave Chinner
2017-09-28  7:41                   ` [PATCH v6 3/4] generic: mmap write readonly DAX file Xiong Zhou
2017-09-28  7:41                   ` [PATCH v6 4/4] generic/413: skip dax to nondax dio test if needed Xiong Zhou
2017-09-29  2:16                   ` [PATCH v7] " Xiong Zhou
2017-09-29 16:24                     ` Ross Zwisler
2017-09-30  2:30                       ` Xiong Zhou
2017-04-11 11:40     ` [PATCH v2 1/3] common: helper to get value from ndctl list by key Eryu Guan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170411142636.au3fnupvd4imuvlx@XZHOUW.usersys.redhat.com \
    --to=xzhou@redhat.com \
    --cc=dan.j.williams@intel.com \
    --cc=eguan@redhat.com \
    --cc=fstests@vger.kernel.org \
    --cc=jmoyer@redhat.com \
    --cc=ross.zwisler@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.